This proposal first aims to improve, develop, and test powerful new molecular techniques to monitor and manipulate dynamic signal transduction in neurons. Goals for voltage indicators based on photoinduced electron transfer include greater sensitivity, genetic targeting, and longer wavelengths. New far-red fluorescent proteins engineered from phycobiliproteins are promising building blocks for in vivo indicators of cell cycle status, Ca2+, and protease activity. A novel alternative approach to measuring and manipulating neuronal activity is to engineer an artificial transcription factor activated by simultaneous high [Ca2+] and illumination, greatly improving on endogenous activity reporters such as c-fos. The genetically encoded snapshot reporter will capture the pattern of activity throughout a large ensemble of neurons at a time precisely defined by the triggering illumination, then drive expression of effector genes to mark those cells and allow selective excitation, inhibition, or ablation to test their functional importance. A chimeric channelrhodopsin activatable by red light permits optogenetic excitation of deep neurons through the intact skull, but its peak wavelength should be further increased and its residual sensitivity to blue light suppressed. Optogenetic inhibition of synaptic release will be improved by a more efficient singlet-oxygen generating protein, and introducing a singlet- oxygen-sensing GFP to map the spatial extent of inhibition. Both optogenetic tools will be applied to dissect amygdalar circuits in fear conditioning. The singlet-oxygen generating protein can now be split into two complementary fragments that only become photoactive after being brought together by chimeric partners. This complementation system may allow protein-protein interactions and kinase and protease activity to be captured for subsequent visualization by electron microscopy. A genetically encoded tag that marks proteins made during a pharmacologically defined period may become applicable to image synthesis and degradation of proteins in intact brain, thanks to development of nanoparticles that deliver small molecule drugs across the blood-brain barrier. Such nanoparticles may also aid clinical drug delivery to the brain. Such techniques will be used to test a new hypothesis that very long-term memories such as fear conditioning are stored as the pattern of holes in the perineuronal net (PNN), a specialized extracellular matrix that envelops mature neurons and restricts synapse formation. The 3-D intertwining of PNN and synapses will be imaged by serial-section electron microscopy. Lifetimes of PNN vs. intrasynaptic components will be compared by pulse-chase 15N labeling in mice and 14C content in human cadaver brains. Genetically encoded indicators and anti-neoepitope antibodies should improve spatial and temporal resolution of the in vivo activity of proteases that locally erode PNN. New techniques including genetic knockouts, better pharmacological inhibitors, and the snapshot reporter should enable more precise inhibition or potentiation of PNN erosion to compare with behavioral consequences. Biosynthesis of PNN components and proteases will be imaged.